My facility is in the process of replacing the fluorescent T8's with Eurilight ET8-1150H's. While doing some energy and RF testing I found that the aluminum heatsink, which runs the entire length of the tube, was carrying voltage.
I have found voltages in various fixtures from ~20VAC all the way to 277VAC (in 277 configuration) and similar results in ballasted configuration. Voltage is being measured between the heatsink and the luminaire's metal frame.
The tubes display the same results whether in opposite end fed or same end fed configurations, typical of a hybrid. The tubes have similar results when put directly into ballasted single or double configurations. I have more testing (sample size) to do on the ballasted ones, but I am quite dismayed by what I am finding so far.
This cannot be right. This is exposed and should be at ground potential. I have never come across a piece of equipment which had (by design) exposed metal at line potential.
I am having an email discussion with someone who is relaying for an engineer. They want video...I am thinking of punctuating the meter test with a lightbulb connected between the heatsink and the frame.
I have found voltages in various fixtures from ~20VAC all the way to 277VAC (in 277 configuration) and similar results in ballasted configuration. Voltage is being measured between the heatsink and the luminaire's metal frame.
The varying voltage does suggest induced voltage
This cannot be right. You betcha !
I am having an email discussion with someone who is relaying for an engineer. They want video...I am thinking of punctuating the meter test with a lightbulb connected between the heatsink and the frame. If it does light up, it's real power .... Post the video here ! :thumbsup:
Comments?
Why is that? Why would a manufacturer/designer do such a thing?
Down here in the U.S., I would start with specifically referring to the NEC :
410.5 Live Parts.
Luminaires, portable luminaires, lampholders, and lamps shall have no live parts normally exposed to contact.
I assume that the CEC has a similar code section.
In one scenario I can imagine, a store owner puts these T8's in their display cases where the customer could easily be shocked by them. It just seems so ludicrous.
I still say it is your meter, cause there is an overcorrect device that is going to be noticing the issue, or else probably some smoke coming out of something or another.
I still say it is your meter, cause there is an overcorrect device that is going to be noticing the issue, or else probably some smoke coming out of something or another.
The heatsink is not bonded, so it's an exposed terminal. It doesn't pop any breakers that way.
When I conducted the test with a light bulb it only worked for a second, then the unintended path in the tube's SMPS went to sh!t in a loud/magic-smoke-release kind of way.
These are just initial findings. I haven't compiled a table for the ballasted fixture tests yet. I am finding similar conditions there, just at the driven frequency from the ballast.
I think you said heat sink wasn't bonded(bad design imho). I guess what you really need to prove is touch potential. FWIW quick fix/ temporary fix run bond jumper?
Why is that? Why would a manufacturer/designer do such a thing?
Down here in the U.S., I would start with specifically referring to the NEC :
410.5 Live Parts.
Luminaires, portable luminaires, lampholders, and lamps shall have no live parts normally exposed to contact.
I assume that the CEC has a similar code section.
In one scenario I can imagine, a store owner puts these T8's in their display cases where the customer could easily be shocked by them. It just seems so ludicrous.
i said that because most of led bulbs doesnt have ground pin so we cant ground them, but the heasinks are supposed to be floating, if you have voltage on it it is defective.
I said that because most of led bulbs doesnt have ground pin so we cant ground them, but the heatsinks are supposed to be floating, if you have voltage on it it is defective.
I think you said heat sink wasn't bonded(bad design imho). I guess what you really need to prove is touch potential. FWIW quick fix/ temporary fix run bond jumper?
This isn't phantom voltage, it's just a bad design.
I will be reverse engineering the SMPS inside the tube, if not for my own curiosity, then to better report it to my safety department as to the root cause. My guess is that the negative side of the power supply's output is supposed to be floating. Since the LED assembly is mounted to an external heat sink it presents an issue.
We have several fixtures which are exhibiting the issue (ballasted and not ballasted). The main one I am focusing on is one which I have the easiest access to and allows quicker testing.
That fixture/luminaire is an Ergolight model 101, made in Canada. The fixture was originally fitted with ballasts. The ballasts (in the case of the 277 VAC/60 Hz stray voltage) have been cut out of the path. The LED tubes are powered directly by 277 via the non-shunted tombstones original to the fixture (rated at 600V).
Even on a test bench with no fixture, the LED tubes exhibit the issue. Depending on which pins are powered (according to the manufacturers instructions) there is voltage present on the heat sink.
The Eurilight LED tubes are of Chinese manufacture, designed by Irtronix in California.
We have several fixtures which are exhibiting the issue (ballasted and not ballasted). The main one I am focusing on is one which I have the easiest access to and allows quicker testing.
That fixture/luminaire is an Ergolight model 101, made in Canada. The fixture was originally fitted with ballasts. The ballasts (in the case of the 277 VAC/60 Hz stray voltage) have been cut out of the path. The LED tubes are powered directly by 277 via the non-shunted tombstones original to the fixture (rated at 600V).
Even on a test bench with no fixture, the LED tubes exhibit the issue. Depending on which pins are powered (according to the manufacturers instructions) there is voltage present on the heat sink.
The Eurilight LED tubes are of Chinese manufacture, designed by Irtronix in California.
Maybe notify HD, that they pose a shock hazard. They for sure sound like a bad design. Maybe they'll remove/discontinue them in the meantime. I'm assuming they're UL listed? Maybe contact UL labs? You might just save a life.
Ok, I didn't understand or missed the part that the heat sink was unbounded to the chassis . In that case holy cow. I would contact Home Depot , that will get the manufacture'r to start paying attention.
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Related Threads
?
?
?
?
?
Electrician Talk
2.3M posts
93.1K members
Since 2007
A forum community dedicated to professional electricians, contractors, and apprentices for residential and commercial work. Come join the discussion about trade knowledge, tools, certifications, wiring, builds, scales, reviews, accessories, classifieds, and more!